COMMUNICATIONS TO THE EDITOR
5792
and acetate, acetate carbon is extensively utilized in the synthesis of glutamic and aspartic acids, leucine and lysine, but not for the synthesis of phenylalanine and tyrosine. Hence neither acetate nor intermediates derived from acetate (aketoglutarate, oxaloacetate) could have been precursors for the benzenoid amino acids. On the other hand, carbon from l-C14 glucose appears in the benzene rings and in the /3 positions of the side chains of the two amino acids (Table 11). Car-
Vol. 52
sole substrate, assuming that under their conditions acetate was utilized by way of glucose. It can be calculated from the data in the tables that the specific activity in the benzene rings was less than that of the glucose. The possibility that the inositol present in the nutrient mediuni is responsible for this isotope dilution is being inl-estigated.
DEPARTMENT OF BIOCHEMISTRY AND THE INSTITUTE OF RADIOBIOLOGY A N D BIOPHYSICS UNIVERSITY OF CHICAGO CHARLES GILVARG4 CHICAGO, ILLINOIS KONRAD BLOCK TABLE I1 RECEIVED XOVEXBER 11, 1950 DISTRIBUTION O F c I 4 I N PK&NYLALANINE AND TYROSINE(4) Public Health Pre-doctoral fellow. IN PROTEIN FROM YEASTGROWN ON 1-C" GLUCOSE ~ l C ~ C a C O O H 5
6
Total COOH" p-Carbonb Picric acid 2,4-Dinitroaniline
TyrosineC r.p.m. Clad
26 * 3 3*l 77 * 8 23 * 8 .
.
I
.
Phenylalanine* c.p.m. C14d
22
=t2 i*l 97 * IO .
.
I
.
.
17 * 2
2 1 1 ..... .. . 37 1 4 C.,8 .... 7$*7 Ca or C6 calcd. a By tyrosine decarboxylase and ninhydrin respectively By decarboxylation of p-hydroxybenzoic acid and of benzoic acid respectively. c Degradations were performed after dilution of the isolated amino acids by non-isotopic phenylalanine and tyrosine. d All samples counted after conversion t o BaC08 and corrected t o infinite thickhess. C1.1,6
bon atoms 1, 3 and 5 of the ring obtained as bromopicrin after conversion of tyrosine to picric acid$ contained no significant radioactivity. Carbons 2 and 6 of the ring, obtained as bromopicrin after conversion of phenylalanine to $bromo, o-nitroacetanilide had a specific activity which accounted for about three-fourths of the total activity in the benzene ring. If two isotopically equilibrated 3-carbon intermediates from l-C1*glucose condensed, any resulting six-membered ring would contain C14 a t adjacent carbon atoms or in para position. The observed isotope distribution rules out this possibility. If, on the other hand, benzene rings were formed by direct cyclization of l-C14 glucose, C14 should be present in one position only. Our data indicate that at least three-fourths of the radioactivity in the benzene rings is present in positions 2 and 6. Since no reactions of glucose other than formation of two-carbon units are known which would lead t o the appearance of CI' in nieta positions, and since acetate is e s cluded as an intermediate i t is likely that only Cz or Ce is labeled. The data are therefore best explained by assuming a cyclization of glucose to form the benzene rings of phenylalanine and tyrosine. This conclusion is in accord with data of fkzddiley, Ehrensvkd, et aZ.,8on tyrosine synthe51s 111 Torulopsis utiLis grown on acetate as the ( ? ) f.Baddiley, G E h r e n s v k d , E Klein, L. Rei0 a n d E.Saluste, J Bid. Ckcirt , 188, 777 (1950)
ADSORPTION OF IRON BY ANION EXCHANGE RESINS FROM HYDROCHLORIC ACID SOLUTIONS*
Sir : As part of a more general study of the anion exchange behavior of metal ions in chloride and fluoride solutions it was found that iron(II1) can be strongly adsorbed from relatively concentrated hydrochloric acid solutions. The results of a series of elution experiments with tracer Fe5yand 0.023 cm.2 columns of Dowex-1 are shown in Table I. The elution constants E = dA/V (where d = distance in cm. an of eluent pass adsorption band moves when V d. through a corumn of cross sectional area A sq. cm.) were found to decrease rapidly with increasing hydrochloric acid concentrations above ca. 1 )If and to reach the very small values of 2 X lo-* in ca. 9 A4 HCI. I n this respect Fe(II1) is quite similar to Pa(V) which was previously discussed.2 Adsorption is probably due to the formation of a strongly adsorbable negatively charged iron complex (or complexes) e.g., FeC14- of probable negative charge minus one, the concentration of which increases with increasing hydrochloric acid concentration. TABLE I ADSORPTION OF Fe(II1) BY DOWEX-1 FROM HYDROCHLORIC ACID SOLUTIONS D 14 HCI E MHCI E Do 0.0148 A7 0 5 1.74 4.0 0 16 1 0 1.34 .0037 270 0 33 5 0 2 0 460 042 1.94 6.0 .00218 0.038 26 9.0 ,00017 5900 3.9 Calculated from equation E = l/(i D) where i is the fractional interstitial space (m. 0.42) and D is the volume distribution coefficient (amount per ml. of resin/ dmount per ml, of solution).
+
I n view of this strong adsorbability of iron it can easily be separated from metal ions which do not form negatively charged complexes (e.g., alkali metals, alkaline earths, etc.). I n this con(1) This document IS based on work performed for t h e Atomic Energy Commission a t t h e Oak Ridge National Laboratory. P a r t of this work was previously reported in report ONRL28G ( J u n r 1949). ( 2 ) K A Kraus and G E. Moore, THIS J o ~ R \ 7 2 , 425J (19.50)
Dec., 1950
COMMUNICATIONS TO THE EDITOR
5793
gives a satisfactory qualitative accounting for the (hkO), (hkl), (hk2), and (hk3) Weissenberg intensities (CuKa), including a prediction of weak (hRO) reflections which are perhaps not altogether inconsistent with those observed. Until intensity work (now in progress) has yielded quantitative data, a choice between this space group and D:d - P3n2 (obtainable with small dis- P4nm tortions of the above structure) or (permitting a closely related structure derived from the “ideal” structure by shifting I1 and IV about 1/4 in z) cannot be made. All other space . groups have been ruled out. The identities of the atoms are not yet known. They may be very difficult to determine because OAKRIDGENATIONAL LABORATORYGEORGE E. MOORE iron and chromium have nearly the same scatterOAK RIDGE,TENNESSEE KURTA. KRAUS ing factors. RECEIVED NOVEMBER 6, 1950 This work is being continued. We are indebted to Professor Pol Duwez and Mr. Paul PietrokowTHE CRYSTAL STRUCTURE OF A SIGMA PHASE, sky of this Institute for the sample of a-FeCr. FeCr’ We are grateful t o Professor Linus Pauling for Sir: helpful discussions, and t o Miss Linda Pauling Despite the importance of the sigma phase in and Mrs. Nan Arp for computational assistance. transition group alloys, the crystal structure of GATESAND CRELLINLABORATORIES OF CHEMISTRY the phase has not heretofore been determined, CALIFORNIA INSTITUTE OF TECHNOLOGY CALIFORNIA DAVIDP. SHOEMAKER primarily because the materials, prepared by PASADENA, No. 1488 BRORGUNNARBERGMAN solid-state transition, are microcrystalline and CONTRIBUTION RECEIVED NOVEMBER 11, 1950 not well suited for single crystal work, while powder photographs have proved too complicated for satisfactory interpretation. SYNTHESIS OF 11-HYDROXYLATED CORTICAL However, we have succeeded in isolating from a STEROIDS. ~~(LY)-HYDROXYCORTICOSTERONE specimen of u-FeCr (46.5 at. yo Cr) two single Sir : crystals roughly 0.1 mm. in size. Single crystal We wish to report the synthesis of 17(a)-hyand powder photography gave a 30-atom primi- droxycorticosterone known as Reichtive tetragonal cell (Laue symmetry D4h), with stein’s Compound Motherwise 1 or Kendall’s Compound ao = 8.799 8.and Ca = 4.546 A. The only observable (hkO) reflections (aside F,2 a substance found by preliminary studies3 to therapeutic activity similar to Cortisone. from a few faint ones a t large Bragg angles) within have The biosynthesis of 17(a)-hydroxycorticosterthe CuKa limit are uniformly strong, and are one from 1l-desoxy-17(a)-hydroxycorticosterone those ({410) , { 3301, { 5501, {720}, { 8201, { 6601, has been demonstrated techniques of perfu{ 9601, { 11.1.01, and { 10.5.0)) which would result sion in the isolated beefusing adrenal gland4 and of infrom fifteen atoms a t the points of a slightly dis- cubation with adrenal homogenates.s torted hexagonal net (with the following typical have synthesized 17(cr)-hydroxycorticoster(x,y)coordinates with respect t o a vertical two- oneWestarting with 20-cyano-17-pregnene-21-01-3,fold axis: I (1) (0, 0); I1 (2) (1/5, 1/5); I11 (2) CHIOR CHzOCOCHs (2/5, 2/5); IV (2) (2/3, 1/31; V (4) (7/15, I I 2/15); VI (4) (11/15, 1/15)) plus fifteen others C-CN CO in a n equivalent net rotated 90’ with respect t o the first. The relative positions of the two nets are as in an “ideal” structure with space group D$ - P4/mnm, as indicated by n-glide extinctions in (Okl) Weissenberg data (CuKa). That “ideal’) structure is ruled out by the general (hkl) intensities. However, atoms IV in the 0 (I) R H (11) second layer have (x,y) coordinates (1/6, 1/6, (Ia) R = CHaCO etc.) very close t o those of I1 in the first layer, and if the eight atoms I1 and IV are moved t o (1) Reichstein, Helo. Chim. Acfa, 80, 953 (1937). (2) Mason, Hoehn and Kendall, J . Biol. Chcm., 114, 459 (1938). new positions S ( j ) with x = 11/60, z = 1/4, a (3) Hench, Kendall, Slocumb and Polley, Arch. I n f . Mcd., 86, 545 structure with space group D$ is obtained which (1950). nection other ions of oxidation number three are of particular interest. Of these, aluminum, chromium and rare earths have been tested to date and they all were found to be practically not adsorbed from strong HC1 solutions. To illustrate the effectiveness of the method an iron impurity (30 mg./l.) was separated from a 2 M aluminum chloride solution in 3 M HC1 using a 10 cm. column of 0.024 sq. cm. cross-section. After passage of 30 ml. of solution (flow rate ca. 0.25 ml. cm. -2 min.-’), when the experiment was interrupted, iron could be detected (visually) only in the first 0.6 cm. of the column. Thus large volumes of solution could be processed with small amounts of resin.
(1) Work done in part under a contract with the Office of Naval Research and in part under a program sponsored by the Carbide and Carbon Chemicals Corporation.
(4) Hechter. Jacobsen, Jeanloz, Levy, Marshall, Pincus and Schenker, Arch. Biochem., 16, 457 (1950). (5) McGinty, Smith, Wilson and Worrd, Science, 111, 506 (1950).
